Effect of Zr doping power on the electrical, optical and structural properties of In–Zr–O anodes

for P3HT : PCBM thin-film organic solar cells

This article has been downloaded from IOPscience. Please scroll down to see the full text article.

2013 J. Phys. D: Appl. Phys. 46 295305

(http://iopscience.iop.org/0022-3727/46/29/295305)

Download details:

IP Address: 163.180.123.209

The article was downloaded on 02/07/2013 at 06:00

Please note that terms and conditions apply.

View the table of contents for this issue, or go to the journal homepage for more

Home Search Collections Journals About Contact us My IOPscience

IOP PUBLISHING JOURNAL OF PHYSICS D: APPLIED PHYSICS

J. Phys. D: Appl. Phys. 46 (2013) 295305 (10pp) doi:10.1088/0022-3727/46/29/295305

Effect of Zr doping power on theelectrical, optical and structuralproperties of In–Zr–O anodes forP3HT : PCBM thin-film organic solar cellsDa-Young Cho1, Kwun-Bum Chung2, Seok-In Na3 and Han-Ki Kim1

1 Department of Advanced Materials Engineering for Information and Electronics, Kyung HeeUniversity, 1 Seocheon-dong, Yongin, Gyeonggi-do 446-701, Republic of Korea2 Department of Physics, Dankook University, Mt. 29, Anseo-Dong, Cheonan 330-714,Republic of Korea3 Professional Graduate School of Flexible and Printable Electronics and Polymer Materials FusionResearch Center, Chonbuk National University, Deokjin-dong, Jellabuk-do561-756, Republic of Korea

E-mail: imdlhkkim@khu.ac.kr

Received 8 April 2013, in final form 19 May 2013Published 1 July 2013Online at stacks.iop.org/JPhysD/46/295305

AbstractWe investigated the effect of Zr doping power on the electrical, optical, structural andmorphological properties of ZrO2 and In2O3 co-sputtered In–Zr–O (IZrO) thin films astransparent anodes for bulk-heterojunction organic solar cells (OSCs). Increased Zr dopingpower led to increased resistivity of as-deposited IZrO films while decreased resistivity ofIZrO films annealed at 500 C. Regardless of the Zr doping power, the IZrO film showed ahigh optical transmittance in the visible wavelength region and near infrared (NIR) wavelengthregion. The optimized IZrO film doped with 50 W ZrO2 radio frequency power showed a sheetresistance of 20.71 /square and transmittance of 83.9%, which is comparable to the value ofconventional In–Sn–O (ITO) films. The electronic structures measured by spectroscopicellipsometry indicated expansion of unoccupied states near the conduction band with increasedZrO2 doping power. OSCs with transparent IZrO anodes exhibited similar performances toOSCs with reference ITO anodes due to the similar optical properties of IZrO and ITO films inthe visible wavelength region. Due to their high transmittance in the NIR wavelength region,IZrO films are a promising replacement for ITO anodes in NIR absorbing OSCs.

(Some figures may appear in colour only in the online journal)

1. Introduction

Rapid advances in organic semiconductor materials and devicefabrication technologies have led to the increased powerconversion efficiency (PCE) of organic solar cells (OSCs)over the last ten years [1–5]. Recently, the PCE values forsmall molecule active layer-based OSCs and polymer activelayer-based OSCs increased up to 12% and 10%, respectively[3, 6]. Most research has focused on synthesizing organicactive layers or hole/electron buffer layers to improve theexciton formation efficiency of OSCs, even though transparentelectrodes are also important components that affect the

performance of OSCs [3, 7]. Because exciton formationefficiency and carrier extraction efficiency are affected by theoptical and electrical properties of transparent electrodes, thePCE of OSCs is closely related to the quality of transparentelectrodes. Several transparent electrode materials have beensuggested as transparent anodes for OSCs, including polymerelectrodes, carbon-based electrodes, metal grids, Ag nanowirenetworks, oxide/metal/oxides multilayer and indium-free TCOelectrodes, such as Al-doped ZnO, Ga-doped ZnO and F-dopedSnO2 [8–17]. Sputtered Sn-doped In2O3 (ITO) films have beenheavily used as transparent electrodes in OSCs and invertedOSCs due to their high conductivity and optical transmittance

0022-3727/13/295305+10$33.00 1 © 2013 IOP Publishing Ltd Printed in the UK & the USA

J. Phys. D: Appl. Phys. 46 (2013) 295305 D-Y Cho et al

[18–24]. However, considering the variety of oxide materialswith a different band gap and optical transmittance, newtransparent conducting oxide (TCO) materials with specialoptical and electrical properties should be designed for OSCsand inverted OSCs to replace ITO anodes. In particular,In2O3-based TCO doped with Mo, Ge, Ti, Si, Ta and Whas been extensively investigated for use in high-mobilityTCO (HMTCO) films due to its high near infrared (NIR)transmittance [25–30]. Among several HMTCO films, Zr-doped In2O3 (IZrO) films are promising TCO materials due totheir high conductivity and optical transmittance in the NIRwavelength region. Kim et al reported that an IZrO filmprepared by pulsed laser deposition (PLD) has a resistivityof 2.5 × 10−4 cm and optical transmittance of 89%, whichis comparable to ITO films [31]. In addition, Koida andKondo suggested that an IZrO film grown epitaxially byPLD is superior to ITO films as a conductive oxide withhigh optical transmittance in the NIR region [32]. Althoughthe electrical and optical properties of IZrO films have beenreported [25, 31–33], there is no report on the application ofIZrO anodes in OSCs. In addition, a detailed investigationregarding the effect of Zr doping power on the resistivity,transmittance, microstructure and electronic structure of IZrOfilms is still lacking even though IZrO films are promisingcandidates for HMTCO films.

In this work, we investigated the effect of Zr dopingpower on the electrical, optical, structural and morphologicalproperties of co-sputtered IZrO films. These films areintended for use as transparent anodes in OSCs instead ofconventional ITO anodes due to their high NIR transparency.By varying the radio frequency (RF) power of the ZrO2 targetduring ZrO2 and In2O3 co-sputtering, we investigated thechanges in the electrical, optical, structural and morphologicalproperties of the as-deposited and annealed IZrO films. Underoptimized conditions, the IZrO film showed a sheet resistanceof 20.71 /square and optical transmittance of 83.9%, whichis comparable to the value of sputtered ITO films. We alsocompared the performances of OSCs with an optimized IZrOanode and a reference ITO anode to show the potential of IZrOanodes for replacing conventional ITO anodes.

2. Experimental details

A dc and RF magnetron sputtering system with dual cathodeswas employed to deposit 200 nm thick Zr-doped In2O3 (IZrO)films on a glass substrate. Using the co-sputtering method,different RF and dc powers were applied to each ZrO2 andIn2O3 target, and IZrO films were prepared as a function ofZrO2 doping power. Before the co-sputtering process, 25 ×25 mm2 glass substrates were cleaned using acetone, methanol,isopropyl alcohol and deionized water in an ultrasonic bathtaking 15 min for each step. They were then dried in nitrogengas. Both ZrO2 (Dasom RMS) and In2O3 (DNT Korea)ceramic targets were pre-sputtered for 20 min under pureAr ambient to remove contaminants on the targets. Afterpre-sputtering ZrO2 and In2O3 targets, the IZrO films weredeposited on the glass substrates at a constant dc power of100 W in the In2O3 target, an Ar flow rate of 10 sccm and

a working pressure of 3 mTorr, as a function of ZrO2 RFdoping power from 0 to 50 W. To obtain uniformly Zr dopedIn2O3 films, the glass substrate was constantly rotated at aspeed of 20 rpm during the ZrO2 and In2O3 co-sputteringprocess. The distance between targets and the glass substratewas maintained at 100 mm. At a constant dc power ofIn2O3, the ZrO2 RF power was varied to investigate theeffect of Zr doping power on the properties of 200 nm thickIZrO films. After deposition of 200 nm-thick IZrO films,all IZrO films were thermally annealed at 500 C under N2

conditions (∼10−1 Torr) for 10 min using a commercial RTAsystem (SNTEK RTA system). The electrical propertiesof the as-deposited and 500 C annealed IZrO films wereanalysed by Hall measurement as a function of Zr dopingpower. The conduction mechanism of the optimized IZrO filmwas investigated based on resistivity change with decreasingtemperature. The resistivity of the optimized IZrO filmwas measured using a physical property measurement system(Quantum Design) in the temperature range from 300 to 5 K.In addition, the optical transmittance of the as-deposited and500 C annealed IZrO films were measured by a UV/visiblespectrometer (UV 540, Unicam). The work function of theoptimized IZrO film was measured by a Kelvin probe (KPTechnology). The structure of the as-deposited and 500 Cannealed IZrO films was examined by x-ray diffraction (XRD:M18XHF-SRA) as a function of Zr doping power. Themicrostructure of the optimized IZrO film was analysed byhigh-resolution transmission electron microscopy (HRTEM:JEM-2100F). Detailed electronic structures of the IZrO films,related to changes in the band gap and unoccupied states in theconduction band, were analysed by spectroscopic ellipsometry(SE). The SE measurement was performed by a rotatinganalyser system with an auto-retarder in the energy range0.75–6.4 eV with incident angles of 65–75.

To apply IZrO films as transparent anodes forOSCs, conventional bulk-heterojunction OSCs based onpoly(3-hexylthiophene) (P3HT, Rieke Metals) and [6,6]-phenyl-C61 butyric acid methyl ester (PCBM, Nano-C)were fabricated. As a reference anode for OSCs,conventional dc-sputtered ITO films were also prepared.The commercial ITO-coated glass was purchased fromSamsung Corning Precision Materials. Before coating witha poly(3,4-ethylenedioxythiophene) : poly(styrenesulfonate)(PEDOT : PSS, Clevios PH510) buffer layer, both IZrO/glassand ITO/glass samples were cleaned by conventional cleaningprocesses. Then, the surfaces of IZrO and ITO anodeswere modified by the UV/ozone treatment for 10 min toimprove wetability of the PEDOT : PSS layer and removethe contaminants on the anodes. PEDOT : PSS buffer layerswere then spin-coated on the UV/ozone-treated IZrO and ITOanodes. Subsequently, the PEDOT : PSS layer was annealed at120 C for 10 min in air. After coating of PEDOT : PSS, a blendof P3HT (50 mg) and PCBM (50 mg) in 1,2-dichlorobenzene(2 ml) was spin-coated onto the PEDOT : PSS layer at 700 rpmfor 60 s in a nitrogen-filled glove box. This was followedby a solvent-annealing treatment for 120 min and additionalthermal annealing at 110 C for 10 min. Finally, a Ca/Al(20/100 nm) cathode with an area of 4.66 mm2 was deposited

2

J. Phys. D: Appl. Phys. 46 (2013) 295305 D-Y Cho et al

on the photoactive layer using a thermal evaporation system.The Ca/Al cathode layer was patterned by a thin shadowmetal mask. All organic and cathode layers on IZrO andreference ITO anodes were prepared by identical coatingprocesses at the same time. The interface in OSCs withIZrO anodes was investigated by HRTEM analysis. Thephotocurrent density–voltage (J–V ) curves of OSCs withIZrO and reference ITO anodes were measured using aKeithley 1200 source measurement unit under 100 mW cm−2

illumination with AM 1.5 G irradiation from a solar simulator.

3. Results and discussion

Figure 1 shows the optical transmittance of the as-depositedand 500 C annealed IZrO films as a function of ZrO2 dopingpower from 10 to 50 W. For comparison, optical transmittanceof a reference ITO film is also shown in figure 1. Althoughthey were grown at room temperature, the as-deposited IZrOfilms shown in figure 1(a) exhibited a typical crystalline In2O3

transmittance with a deep valley in the 600 nm wavelengthregion regardless of the ZrO2 doping power. Due to theexistence of nanocrystalline In2O3 or Zr-doped In2O3 phase,the as-deposited IZrO films showed optical transmittanceunlike the amorphous In2O3 phase [24]. The upper panelpictures show the transparency of the as-deposited IZrO filmswith increasing ZrO2 doping power. Below ZrO2 power of30 W, all as-deposited IZrO films showed similar transparencywith a greenish colour. However, the IZrO films doped atZrO2 powers of 40 and 50 W exhibited a slightly bluish colour.The IZrO films annealed at 500 C also showed high opticaltransmittance regardless of the ZrO2 doping power, as shownin figure 1(b). Although they underwent annealing at 500 C,they showed similar optical transmittance to the as-depositedIZrO films. In addition, the sample colour in the upper panel issimilar to that of the as-deposited IZrO films. At a wavelengthof 550 nm, an increase in ZrO2 doping power from 10 to 50 Wdecreased the optical transmittance from 82.66% to 76.12%. Inaddition, an increase in ZrO2 doping power led to high energyshifts of the band edges of the IZrO films. High energy shiftsof band edges indicate that the band gaps of the IZrO filmsincrease with increasing ZrO2 doping power. The widenedband gap that results from increased ZrO2 doping power isrelated to the Burstein–Moss (BM) effect [34]. When In2O3

is doped with Zr, the conduction band of IZrO is partiallyfilled with electrons, and the Fermi level is located inside theconduction band. Filling electrons in the conduction band canshift the Fermi level to a higher energy state in the conductionband. Therefore, transitions to lower energy are blocked,which is known as the BM effect. In addition, the transparencyof the IZrO films in the NIR wavelength region above 800 nmwas much higher than that of the conventional ITO films andother HMTCO films [26–30]. Note that there was no change intransparency of NIR region, indicating that the annealed IZrOfilms possess similar free carrier concentrations regardless ofthe ZrO2 power.

Figure 2 shows the Hall measurement results of the as-deposited IZrO films as a function of the ZrO2 doping power.The 200 nm thick as-deposited IZrO films shown in figure 2(a)

Figure 1. (a) Optical transmittance of the as-deposited and(b) 500 C annealed IZrO films as a function of the ZrO2 RF power.The upper panel exhibits the transparency of the IZrO films withincreasing Zr doping power.

showed a linear increase in resistivity and sheet resistancewith increasing ZrO2 doping power from 10 to 50 W. The10 W Zr-doped In2O3 (IZrO) film showed a resistivity of1.5 × 10−3 cm, while the 50 W Zr-doped In2O3 (IZrO)film showed an increased resistivity of 3.19 × 10−3 cm.The increased resistivity of the IZrO film with increasedZrO2 doping power is attributed to the insulating propertiesof the co-sputtered ZrO2 dopant in the In2O3 matrix. Theactivation energy for substituting Zr4+ dopant with In3+ atomsto provide excess electrons is insufficient at room temperature.Figure 2(b) shows the carrier mobility and concentrations of theas-deposited IZrO films with increasing ZrO2 doping power.The decrease in carrier concentration with increased ZrO2

doping power indicates that Zr atoms cannot act as a dopant at

3

J. Phys. D: Appl. Phys. 46 (2013) 295305 D-Y Cho et al

10 20 30 40 500

40

80

120

160

200 Sheet resistance Resistivity

ZrO2 RF power [W]

She

et r

esis

tanc

e [Ω

/squ

are]

2.0x10-3

4.0x10-3

6.0x10-3

8.0x10-3

1.0x10-2

Res

istiv

ity [Ω

cm

]

10 20 30 40 500

10

20

30

40

50

60

Mob

ility

[cm

2 V–1

s–1]

ZrO2 RF power [W]

1019

1020

1021

Mobility Carrier Concentration

Car

rier

conc

entr

atio

n[c

m–3

](b)

(a)

Figure 2. (a) Sheet resistance, resistivity, (b) mobility and carrierconcentration of the as-deposited IZrO films with increasing ZrO2

RF power from 10 to 50 W.

room temperature because they cannot effectively substituteat In sites. Because Zr atoms were doped into the In2O3

matrix by sputtering with a ZrO2 target rather than a metallicZr target, significantly higher activation energy was necessaryto decrease the resistivity of the IZrO films. Therefore, a Zrdopant activation process that uses rapid thermal annealing(RTA) is imperative to obtain low-resistivity IZrO films.

Figure 3(a) shows the resistivity and sheet resistance ofthe 500 C annealed IZrO films with increased ZrO2 dopingpower. Unlike the as-deposited films, the 500 C annealedIZrO films showed decreased resistivity and sheet resistancewith increased ZrO2 doping power. At 50 W Zr dopingpower, the IZrO film showed the lowest resistivity of 4.14 ×10−4 cm and sheet resistance of 20.71 /square. The lowresistivity of the 500 C annealed IZrO films is attributed tothe efficient activation of Zr dopant and crystallization ofIZrO films. Like Sn4+ dopants in ITO films, a Zr4+ ion cansubstitute the In3+ site and produce an excess electron inthe In2O3 matrix after RTA. The high carrier concentrationof 4.11 × 1020 cm−3 and mobility of 36.7 cm2 V−1 s−1 of theIZrO films prepared at 50 W ZrO2 power confirmed that theZr dopant was effectively activated. In addition, the RTAprocess improved the crystallinity of the IZrO films. ForITO films, a high carrier concentration caused by high Sndopant contents above 10 wt% is the main factor causing lowresistivity. However, low resistivity was obtained with lowZr dopant contents for IZrO films due to fairly high carriermobility [32]. Therefore, the low resistivity of the annealedIZrO films is attributed to increased carrier mobility, as shown

10 20 30 40 500

20

40

60

80

100 Sheet resistance Resistivity

ZrO2 RF power [W]

She

et r

esis

tanc

e [Ω

/squ

are]

2.0x10-4

4.0x10-4

6.0x10-4

8.0x10-4

1.0x10-3

1.2x10-3

Res

istiv

ity [Ω

cm

]

10 20 30 40 5010

20

30

40

50

60

Mob

ility

[cm

2 V–1

s–1]

ZrO2 RF power [W]

1020

1021

Mobility Carrier Concentration

Car

rier

conc

entr

atio

n [c

m–3

]

(a)

(b)

Figure 3. (a) Sheet resistance, resistivity, (b) mobility and carrierconcentration of the 500 C annealed IZrO films with increasingZrO2 RF power from 10 to 50 W.

1 10 1000

1

2

3

4

5

6

7HeatingCooling

Heating

Cooling

Temperature [K]

Res

istiv

ity [1

0-4Ω

cm

]

500oC Annealed IZrO

Figure 4. Temperature dependence of the resistivity of theoptimized IZrO film measured from 5 to 300 K.

in figure 3(b). Compared with the previously reported IZrOfilms prepared by PLD [31, 32], the IZrO films prepared bysputtering showed lower carrier mobility because Zr activationwas carried out by a post-annealing process.

To understand the conduction mechanism of the optimizedIZrO films, the resistivity was measured as a function oftemperature. Figure 4 presents the temperature dependence ofthe resistivity of the optimized IZrO films (50 W ZrO2 power)as a function of temperature from 5 to 300 K. It was clearlyshown that the resistivity of the IZrO film decreased withdecreasing temperature. Positive dependence (dρ/dT > 0)

of resistivity on temperature indicates that the IZrO filmhad typical metallic conduction behaviour due to degenerated

4

J. Phys. D: Appl. Phys. 46 (2013) 295305 D-Y Cho et al

10 20 30 40 500

20

40

60

80

100

ZrO2 RF power [W]

Tra

nsm

ittan

ce [%

]

0

2

4

6

8

Transmittance at 550 nm Figure of merit

Fig

ure

of m

erit

[10-3

Ω-1]

10 20 30 40 5060

70

80

90

100

ZrO2 RF power [W]

Tra

nsm

ittan

ce [%

]

2

4

6

8

Transmittance at 400 nm to 1200 nm Figure of merit F

igur

e of

mer

it [1

0-3Ω

-1]

(a)

(b)

Figure 5. Figure of merit values for IZrO films calculated from thesheet resistance (Rsh) and optical transmittance (T ) of the IZrOfilms at (a) 550 nm wavelength, and (b) average opticaltransmittance between 400 and 1200 nm.

semiconductor properties. Similar to our previously reportedHMTCO films [26–30], the resistivity of the IZrO filmsdecreased with decreasing temperature because the carrierelectron flow was dominated by an electron–phonon scatteringmechanism. As the temperature decreased, electron–phononscattering decreased, resulting in a decrease in the IZrOresistivity.

Figure 5 shows the figure of merit (T 10/Rsh) valuesobtained from the sheet resistance (Rsh) and opticaltransmittance (T ) of the 500 C annealed IZrO films as afunction of the Zr doping power. The figure of merit value wascalculated using optical transmittance at 550 nm and averagetransmittance between 400 and 1200 nm wavelengths. Assuggested by Haacke [35], the figure of merit values arewell-known criteria for deciding the optimized conditions ofTCO films. They are also used to compare the quality of TCOfilms. Figure 5(a) exhibits the figure of merit values calculatedfrom the transmittance of IZrO films at a wavelength of 550 nm.Due to the slightly lower optical transmittance of the IZrOfilm at 550 nm, the IZrO film prepared at 50 W doping powerhad a lower figure of merit value than those of the IZrO filmsprepared at 20–40 W doping power. The figure of merit valuecalculated from average transmittance, shown in figure 5(b),increased with increasing ZrO2 doping power due to decreasedsheet resistance, as expected from figure 3(a). The IZrO filmprepared at a ZrO2 doping power of 50 W showed the highest

Table 1. Comparison of the 50 W IZrO film and reference ITO filmfor OSCs.

Properties IZrO (50 W) Ref.-ITO

Sheet resistance (/square) 20.71 14.55Mobility (cm2 V−1 s−1) 36.70 28.22Carrier concentration (cm−3) 4.11 × 1020 1.5 × 1021

Transmittance at 550 nm wavelength (%) 76.12 79.40Work function (eV) 5.26 4.80RMS roughness (nm) 1.97 1.25

20 30 40 50 60 70 80

(222

)

Inte

nsity

[a.u

.]2 θ [°]

50W

40W

30W

20W

10W

(400

)

As-deposited IZrO

20 30 40 50 60 70 80

(622

)

(440

)

(400

)(222

)

Inte

nsity

[a.u

.]

2 θ [°]

50W

40W

30W

20W

10W

500oC Annealed IZrO

(a)

(b)

Figure 6. XRD plots of the (a) as-deposited and (b) 500 Cannealed IZrO films as a function of the ZrO2 doping power.

figure of merit value of 8.37×10−3 −1, which is comparableto that of conventional ITO films (6.8×10−3 −1). Therefore,the optimized doping power of ZrO2 was determined to be50 W. Table 1 summarizes the properties of the optimized IZrOand reference ITO electrodes.

Figure 6(a) shows the XRD plots of the as-depositedIZrO films with increasing ZrO2 doping power. Regardlessof the ZrO2 doping power, all XRD plots of the as-depositedIZrO films exhibited two broad (2 2 2) and (4 0 0) peaks at2θ = 29.7 and 34.56. Because all films were preparedat room temperature, the as-deposited IZrO film consisted ofnanocrystalline In2O3 or Zr-doped In2O3 phase with (2 2 2) and(0 0 4) preferred orientation. Like conventional crystalline ITOfilms, the XRD plot of the 500 C annealed IZrO films showedseveral strong peaks including (2 2 2), (4 0 0), (4 4 0) and (6 2 2)at 2θ = 30.7, 35.54, 51.18 and 60.94, respectively.Because the ion radius of Zr4+ (0.72 Å) is similar to that of

5

J. Phys. D: Appl. Phys. 46 (2013) 295305 D-Y Cho et al

Figure 7. (a) Absorption coefficient of the IZrO films as a functionof the ZrO2 doping power. (b) Schematic electronic orbital structureof In2O3 and InZrO films.

In3+ ions (0.80 Å), the annealed IZrO film showed a bixbyitestructure similar to that of ITO films [32]. In particular, thestronger intensity of the (2 2 2) peak of the annealed IZrO filmsindicates that the 500 C annealed IZrO film has strongly (2 2 2)preferred grains regardless of the ZrO2 doping power.

To investigate the electronic structure of the IZrO filmsnear the conduction band, SE measurements were carried outas a function of the ZrO target plasma power. The absorptioncoefficient spectra of the IZrO films on Si substrates are shownin figure 7(a). These spectra were extracted from a simplefour-phase model, comprised of a Si substrate, SiO2 overlayer,IZrO overlayer and an ambient layer [36]. The extractedoptical band gap slightly increased with increased ZrO2 dopingpower. This implies an increase in incorporated ZrO2 withthe band gap of ∼5 eV, which is similar to the transmittanceresults shown in figure 1. Interestingly, appearance of theonset energy above the optical band gap around ∼3.5 eVincludes the conduction band states of IzrO electrodes. Thesecorrespond to hybridized molecular orbital states mixed by theIn 5sp + O 2p states and Zr 4d + O 2p states [29, 37]. Theconduction band near the band gap of the IZrO film with 50 WZrO2 doping power is significantly expanded by increasing theincorporated ZrO2. This indicates an increase in unoccupiedabsorption states caused by adding the hybridized states ofZr 4d + O 2p into the matrix of In 5sp + O 2p states, whichimproves charge transport. The schematic diagram plotted infigure 7(b) indicates increases in the number of unoccupiedstates. As a result, the changes in electronic structure causedby incorporating ZrO2 are strongly related to the low resistivityof the InZrO films with 50 W ZrO2 doping power.

AFM analysis was employed to compare the surfacemorphology of the as-deposited and 500 C annealed IZrOfilms prepared at 50 W ZrO2 doping power. Figure 8(a)exhibits a surface AFM image of the as-deposited IZrO film,

Figure 8. Surface AFM images of the (a) as-deposited and(b) 500 C annealed IZrO films prepared at a ZrO2 doping power of50 W.

which has very smooth surface morphology and a low rootmean square (RMS) roughness of 2.02 nm. Although the IZrOsample was prepared at room temperature, the surface AFMimage of the as-deposited sample showed crystalline features,which is consistent with XRD results. The 500 C annealedIZrO film in figure 8(b) also has a very smooth surface eventhough it has (2 2 2) and (4 0 0) preferred orientation after theRTA process. Unlike conventional ITO films grown by insitu substrate heating, the IZrO films were crystallized byirradiation heating with a halogen lamp in the RTA system[38]. Because of this, the IZrO films showed fairly smoothsurface morphology. The RMS roughness of the annealedIZrO films was 1.97 nm, which is similar to that of the as-deposited IZrO films. The smooth and featureless surface ofthe transparent anode without protrusion, cracks and holesis one of the important factors affecting the printing-basedcoating process of an organic active layer. In general, organiclayers are directly coated on a transparent electrode usingvarious printing processes. Therefore, the smoothness of thetransparent electrode critically affected the uniformity of theorganic layer. In addition, the shunt resistance of OSCs maybe affected by the surface morphology of transparent anodes.Therefore, the smooth surface of the IZrO films possessed

6

J. Phys. D: Appl. Phys. 46 (2013) 295305 D-Y Cho et al

460 455 450 445 4400

50

100

150

200

250

300

In 444.4 eV

451.9 eV

Inte

nsity

[103 C

ount

s/s]

Binding energy [eV]

194 192 190 188 186 184 182 180 178 176 1745.0

5.5

6.0

6.5

7.0

Zr

184.2 eV181.8 eV

Inte

nsity

[103 C

ount

s/s]

Binding energy [eV]

0 1000 20000

20

40

60

80

100

Ato

mic

per

cent

[%]

Etch time [s]

In Zr O

545 540 535 530 52520

40

60

80 O

531.13 eV

529.9 eV

Inte

nsity

[103 C

ount

s/s]

Binding energy [eV]

(a) (b)

(c) (d)

Figure 9. (a) XPS depth profile of the optimized IZrO film prepared at a ZrO2 doping power of 50 W. Core-level spectra of (b) In 3d, (c) Zr3d and (d) O 1s obtained from the optimized IZrO film.

surface properties that are desirable for fabrication of printing-based OSCs [39].

Figure 9(a) shows the x-ray photoelectron spectroscopy(XPS) depth profile of the optimized IZrO film prepared at aZrO2 doping power of 50 W. The XPS depth profile showsconstant In, Zr and O atomic percentages with increasedetching time, indicating that the co-sputtered IZrO film has auniform composition. The Zr content in the IZrO film was verylow (∼0.5 at%) even though it was prepared at 50 W RF power.The low Zr content led to high NIR transmittance because thefree carrier concentration in the TCO films is critically relatedto the NIR transmittance [41]. The detailed binding energy ofthe IZrO film was analysed by the core-level spectra of In 3d, Zr3d and O 1s, as shown in figures 9(b)–(d). The In, Zr and Obinding energies are located at 451.93 eV (In 3d3/2), 444.43 eV(In 3d5/2), 184.23 eV (Zr 3d3/2) and 181.83 eV (Zn 3d5/2),respectively. Peak positions matched well with the core-levelspectra of previously reported IZrO films [40]. However, theO 1s peak shows multiple components at binding energies of531.13 and 529.98 eV resulting from oxygen vacancies (OI )

and O–In binding (OII ). The presence of two types of O2 ionsin the IZrO film caused the two levels (OI and OII ) of bindingenergy. It has been reported that the lower binding energy peak(OII ) is from O2 ions that neighbour indium atoms with theirfull complement of electrons, while the higher binding energypeak (OI ) corresponds to oxygen vacancies [41].

HRTEM analysis was carried out to investigate themicrostructure of the optimized IZrO film. Figure 10(a) showsa cross-sectional HRTEM image obtained from the 500 Cannealed IZrO film prepared at 50 W ZrO2 doping power.

As expected from the XRD results in figure 6, the IZrO filmpossessed a well-developed columnar structure consisting ofsubgrains with a strongly preferred (2 2 2) orientation. TheHRTEM image also shows that the surface of the 500 Cannealed IZrO film is very smooth, as confirmed by AFManalysis. Although the IZrO film was annealed at 500 C,the formation of a columnar structure does not affect themorphology of the IZrO films. Figure 10(b) shows enlargedHRTEM images and fast Fourier transform (FFT) patternsobtained from an IZrO film from the bottom (A′) and middle(B′) regions. The letters in figure 10(b) indicate the HRTEMimages and FFT patterns that were obtained from the samelettered region in the cross-sectional image in figure 10(a).The enlarged HRTEM image obtained from the middle region(B′) of the IZrO film has a well-developed bixbyite structurewhen compared with the HRTEM image obtained from thebottom region (A′), which has characteristics of a columnarstructure. The strong spot in the FFT pattern (B′) indicatedthat the crystallinity of the IZrO film improved with increasingthickness. Because the halogen lamp in the RTA processeffectively heats at the surface of the IZrO films, the IZrOfilm showed better crystallinity at the surface region than atthe bottom region, as shown in figure 10(b).

To investigate the feasibility of using IZrO films astransparent anodes for OSCs, the performance of OSCscomposed of a P3HT : PCBM active layer was examined.Figure 11(a) shows a picture of bulk-heterojunction OSCsfabricated on the optimized IZrO anode. Figure 11(b) shows across-sectional TEM image obtained from the OSC fabricatedon an optimized IZrO anode. The cross-sectional TEM

7

J. Phys. D: Appl. Phys. 46 (2013) 295305 D-Y Cho et al

(a)

(b)

Figure 10. (a) Cross-sectional HRTEM image of the optimizedIZrO film. (b) Enlarged HRTEM images and FFT patterns obtainedfrom the indicated alphabet region (A′ and B′) in the cross-sectionalHRTEM image.

image clearly shows a well-defined IZrO anode, PEDOT : PSS,P3HT : PCBM active layer and Al : Ca cathode layer withoutinterfacial layers. The sharp interface between the IZrO andorganic layers suggests that the IZrO anode is chemically stableagainst the acidic PEDOT : PSS solution. Figure 11(c) showsenlarged HRTEM images obtained from the interface betweenthe IZrO and acidic PEDOT : PSS layers. The top region of theIZrO anode also consisted of well-developed crystallites with(2 2 2) preferred orientation, similar to the middle region shownin figure 10(b). The enlarged lattice image in figure 11(c)shows that the IZrO film has a lattice constant of 2.885 Å, whichis similar to the value calculated from XRD results (2.908 Å).

Photocurrent density–voltage (J–V ) curves for the OSCsfabricated on reference ITO and IZrO anodes are shown infigure 12(a), which includes an inset picture of the OSC withan IZrO anode. The photovoltaic characteristics of OSCs weremeasured under 100 mW cm−2 illumination and AM 1.5 Gconditions. Table 2 presents cell performances such as thefill factor (FF), short-circuit current density (Jsc), open-circuitvoltage (Voc) and PCE of OSCs with different anodes. Asdemonstrated by figure 12(a), the performance characteristicsof OSCs fabricated on IZrO 1 and IZrO 2 anodes were obtainedas follows: FF, 61.71% and 62.14%; Jsc, 8.484 mA cm−2

and 8.293 mA cm−2; Voc, 0.593 V and 0.597 V; PCE, 3.106%and 3.077%. The cell performances of OSCs with an IZrOelectrode were slightly inferior to those of the OSCs with areference ITO anode with FF of 65.03%, Jsc of 8.833 mA cm−2,Voc of 0.608 V and PCE of 3.495%. Although the IZrOfilm has a similar sheet resistance and optical transmittancewith reference ITO, the OSCs with the IZrO anode show

(a)

(b)

(c)

Figure 11. (a) Picture of OSCs fabricated on a transparent IZrOanode. (b) Cross-sectional TEM image of OSCs fabricated on anIZrO anode (c) Enlarged interface between the IZrO anode and thePEDOT : PSS layer. The enlarged top region shows that the grainshowed a (2 2 2) preferred orientation.

a 10% lower PCE value than the reference sample due tothe lower FF value and current density. Because the OSCfabrication process on the IZrO anode was not optimizedunlike the reference ITO anode, further improvement of deviceperformance is necessary, The energy level band diagramof OSCs with an IZrO anode is illustrated in figure 12(b).The work function of IZrO (5.264 eV) and ITO (4.80 eV)was investigated with a Kelvin probe measurement, and eachenergy level of PEDOT : PSS, P3HT and PCBM was obtainedfrom the literature [42]. Although the work function of theIZrO anode was different from that of the ITO anode, their Voc

values were similar due to the identical PEDOT : PSS bufferlayer. This allows ohmic contact between the anode and theactive layer [18]. It is thought that sheet resistance and opticaltransmittance have greater effects on the device performanceof OSCs than the work function of the anode [43]. Thus,

8

J. Phys. D: Appl. Phys. 46 (2013) 295305 D-Y Cho et al

Figure 12. (a) Current density (J )–voltage (V ) characteristics ofOSCs fabricated on reference ITO and optimized IZrO anodes (IZrO1 and IZrO 2). The inset shows an OSC with different IZrO 1 andIZrO 2 anodes. (b) Band diagram of a P3HT : PCBM based OSCwith an IZrO anode.

Table 2. Cell performances of OSCs fabricated on reference ITOand 500 C annealed IZrO anodes. Both IZrO1 and IZrO2 led to thesame OSC with different cathode positions, as indicated by arrowsin the inset of figure 12(a).

FF (%) Jsc (mA cm−2) Voc (V) PCE (%)

Reference ITO 65.03 8.833 0.608 3.495IZrO 1 61.71 8.484 0.593 3.106IZrO 2 62.14 8.293 0.597 3.077

the excellent cell performances of IZrO-based OSCs may belargely due to the low sheet resistance of 20.71 /square andhigh transmittance of 83.9%. In addition, the OSCs fabricatedon IZrO electrodes utilized P3HT : PCBM active layers thatabsorb light wavelength between 400 and 650 nm. Improvedcell performance is expected, if an organic active layer thatabsorbs wavelengths from the NIR region is employed. Thisis due to the high transmittance of the IZrO films in the NIRregion. Therefore, it is believed that IZrO films are a promisingTCO electrode material for OSCs or tandem OSCs with activematerials that absorb the NIR region of light.

4. Conclusion

We investigated the effect of Zr doping power on the propertiesof IZrO films prepared by co-sputtered ZrO2 and In2O3 targetsas an alternative to ITO anodes for OSCs. At ZrO2 dopingRF power of 50 W and rapid thermal annealing temperature

of 500 C, we obtained a high-quality IZrO anode with alow resistance of 20.71 /square, high transmittance of 83.9%and work function of 5.26 eV. Due to the similar ion radii ofZr4+ dopant and In3+ ions, the IZrO films showed a bixbyitestructure similar to that of In2O3 films regardless of the Zrdoping power. Based on SE analysis, we suggested thepossible electronic structure of the IZrO films and the effectof the ZrO2 doping power. Furthermore, the IZrO filmsexhibited smooth and featureless surfaces even though theyshowed (2 2 2) preferred orientation. Using an optimized IZrOanode, we fabricated a high-performance OSC with a PCE of3.106%, which is comparable to that of an OSC fabricated on acommercial ITO anode (3.495%). This result indicates that theIZrO film is a high-quality In2O3-based TCO anode materialwith high transmittance in the NIR wavelength region. IZrOfilms are promising substitutes for conventional ITO anodes inOSCs that absorb the NIR region of light.

Acknowledgments

This work was supported by the Industrial Strategic technologydevelopment program (no 10042412), funded by the Ministryof Knowledge Economy (MKE, Korea) and partially supportedby the New and Renewable Energy Program of the KoreaInstitute of Energy Technology Evaluation and Planning(KETEP) funded by the Ministry of Knowledge Economy(MKE) (20113010010030).

References

[1] Jørgensen M, Norrman K, Gevorgyan S A, Tromholt T,Andreasen B and Krebs F C 2012 Adv. Mater. 24 580

[2] Krebs F C 2009 Sol. Energy Mater. Sol. Cells 93 394[3] Ameri T, Dennler G, Lungenschmied C and Brabec C J 2009

Energy Environ. Sci. 2 347[4] Li G, Zhu R and Yang Y 2012 Nature Photon. 6 153[5] Service R F 2011 Science 332 293[6] Pfeiffer M, Bauerle P and Leo K 2013 Heliatek Co. report

(Heliatek, Dresden, Germany)[7] Po R, Carbonera C, Bernardi A and Camaioni N 2011 Energy

Environ. Sci. 4 285[8] Lim J-W, Cho D-Y, Kim J, Na S-I and Kim H-K 2012 Sol.

Energy Mater. Sol. Cells 107 348[9] Lee J-H, Shin H-S, Noh Y-J, Na S-I and Kim H-K 2013 Sol.

Energy Mater. Sol. Cells 114 15[10] Lyons S De P E, Sorel S, Doherty E M, King P J, Blau W J,

Nirmalraj P N, Boland J J, Scardaci V, Joimel J andColeman J N 2009 ACS Nano 3 714

[11] Kang M-G, Kim M-S, Kim J and Guo L J 2008 Adv. Mater.20 4408

[12] Cho S-W et al 2008 Thin Solid Films 516 7881[13] Jeong J-A and Kim H-K 2009 Sol. Energy Mater. Sol. Cells

93 1801[14] Park J-H, Ahn K-J, Na S-I and Kim H-K 2011 Sol. Energy

Mater. Sol. Cells 95 657[15] Murdoch G B, Hinds S, Sargent E H, Tsang S W,

Mordoukhosvski L and Lu Z H 2009 Appl. Phys. Lett.94 203301

[16] Hu Z, Zhang J, Hao Z, Hao Q, Geng X and Zhao Y 2011 Appl.Phys. Lett. 98 123302

[17] Yoshida Y, Tanaka S, Hiromitsu I, Fujita Y and Yoshino K2008 Japan. J. Appl. Phys. 47 867

9

J. Phys. D: Appl. Phys. 46 (2013) 295305 D-Y Cho et al

[18] Yeo J-S, Yun J-M, Kim D-Y, Park S, Kim S-S, Yoon M-H,Kim T-W and Na S-I 2012 ACS Appl. Mater. Interfaces24 2551

[19] Cho C-K, Hwang W-J, Eun K, Choa S-H, Na S-I and Kim H-K2011 Sol. Energy Mater. Sol. Cells 95 3269

[20] Barnes T M, Bergeson J D, Tenent R C, Larsen B A, Teeter G,Jones K M, Blackburn J L and Lagemaat J 2010 Appl. Phys.Lett. 96 243309

[21] Choi Y-Y, Kang S J, Kim H-K, Choi W M and Na S I 2012Sol. Energy Mater. Sol. Cells 96 281

[22] Lee J-Y, Connor S T, Cui Y and Peumans P 2010 Nano Lett.10 1276

[23] Kang M-G, Xu T, Park H J, Luo X and Guo L J 2010 Adv.Mater. 22 4378

[24] Choi K-H, Jeong J-A, Kang J-W, Kim D-G, Kim J K, Na S-I,Kim D-Y, Kim S-S and Kim H-K 2009 Sol. Energy Mater.Sol. Cells 93 1248

[25] Calnan S and Tiwari A N 2010 Thin Solid Films 518 1839[26] Shin Y-H, Na S-I and Kim H-K 2012 J. Vac. Sci. Technol. A

30 061507[27] Kang S-B, Lim J-W, Na S-I and Kim H-K 2012 Sol. Energy

Mater. Sol. Cells 107 373[28] Lim J-W, Na S-I and Kim H-K 2012 Sol. Energy Mater. Sol.

Cells 98 409[29] Lee H-M, Kang S-B, Chung K-B and Kim H-K 2013 Appl.

Phys. Lett. 102 021914

[30] Kim J H, Shin Y-H, Seong T-Y, Na S-I and Kim H-K 2012J. Phys. D: Appl. Phys. 45 395104

[31] Kim H, Horwitz J S, Kushto G P, Qadri S B, Kafafi Z Hand Chrisey D B 2001 Appl. Phys. Lett. 78 1050

[32] Koida T and Kondo M 2006 Appl. Phys. Lett.89 082104

[33] Asikainen T, Ritala M and Leskela M 2003 ThinSolid Films 440 152

[34] Burstein E 1954 Phys. Rev. 93 632[35] Haacke G 1976 J. Appl. Phys. 47 4086[36] Park H-W, Park J-S, Lee J H and Chung K-B 2012

Electrochem. Solid-State Lett. 15 H133[37] Richter J H, Karlsson P G, Sanyal B, Blomquist J, Uvdal P and

Sandell A 2007 J. Appl. Phys. 101 104120[38] Choi H-J, Yoon S-G, Lee J-H and Lee J-Y 2012 ECS J. Solid

State Sci. Technol. 1 Q106[39] Chauhan R N, Singh C, Anand R S and Kumar J 2012 Int. J.

Photoenergy 2012 1[40] Ginley D S, Hosono H and Paine D C 2010 Handbook of

Transparent Conductors (Berlin: Springer)[41] Wen S J, Campet G, Portier J, Couturier G and

Goodenough J B 1992 Mater. Sci. Eng. B 14 115[42] Arco L G D, Zhang Y, Schlenker C W, Ryu K, Thompson M E

and Zhou C 2010 ACS Nano 4 2865[43] Pandey A K, Nunzi J M, Ratier B and Moliton A 2008 Phys.

Lett. A 372 1333

10